With a welder and a bunch of scrap, you can build just about anything that moves. Want a dune buggy? That’s just some tube and a pipe bender. Need a water pump? You might need a grinder. A small tractor? Just find some big knobby tires in a junkyard. Of course, the one thing left out of all these builds is a small motor, preferably one that can run on everything from kerosene to used cooking oil. This is the problem [Shane] is tackling for his entry to the 2016 Hackaday Prize. It’s an Open Source Two-Stroke Diesel Engine that’s easy for anyone to build and has minimal moving parts.
[Shane]’s engine is based on the Junkers Jumo 205 motor, a highly successful aircraft engine first produced in the early 1930s and continued production through World War II. This is a weird engine, with two opposed pistons in one cylinder that come very close to slamming together. It’s a great design for aircraft engines due to it’s lightweight construction. And the simplicity of the system lends itself easily to wartime field maintenance.
The Jumo 205 was a monstrous 12-piston, 6-cylinder engine, but for [Shane]’s first attempt, he’s scaling the design down to a 50cc motor with the intent of scaling the design up to 125cc and 250cc. So far, [Shane] has about 30 hours of simple CAD work behind him and a ton of high-level FEA work ahead of him. Then [Shane] will actually need to build a prototype.
This is actually [Shane]’s second entry to the Hackaday Prize with this idea. Last year, he threw his hat into the ring with the same idea, but building a working diesel power plant is a lot of work. Too much for one man-year, certainly, so we can’t wait to see the progress [Shane] makes this year.
Electric cars are all the rage lately, but let’s not forget about the old standby – internal combustion. The modern internal combustion engine is a marvel of engineering. Today’s engines and surrounding systems have better power, greater fuel economy, and lower emissions than anything that has come before. Centuries’ worth of engineering hours have gone into improving every aspect of the engine – with one notable exception. No automotive manufacturer has been able to eliminate the engine’s camshaft in a piston powered-production vehicle. The irony here is that camless engines are relatively easy to build. The average hacker could modify a small four-stroke engine for camless operation in their workshop. While it wouldn’t be a practical device, it would be a great test bed for experimentation and learning.
Suck, Squeeze, Bang, Blow
A multi-cylinder gasoline engine is a complex dance. Hundreds of parts must move in synchronicity. Valves open and close, injectors mist fuel, spark plugs fire, and pistons move up and down. All follow the four-stroke “Intake, Compression, Combustion, Exhaust” Otto cycle. The camshaft controls much of this by opening and closing the engine’s spring-loaded intake and exhaust valves. Lobes on the shaft press on tappets which then move the valve stems and the valves themselves. The camshaft itself is driven at half the speed of the crankshaft through timing gears, chains, or a belt. Some valve trains are relatively simple – such as overhead cam engines. Others, such as the cam-in-block design, are more complex, with pushrods, rockers, and other parts required to translate the movement of the cam lobe to movement at the valve.
Exactly when, and how fast a valve opens is determined by the profile of the cam lobe. Auto racing and performance enthusiasts often change camshafts to those with more aggressive profiles and different timing offsets depending on the engine’s requirements. Everything comes at a cost though. A camshaft machined for maximum power generally won’t idle well and will make the engine harder to start. Too aggressive a lobe profile can lead to valve float, where the valves never fully seat at high RPM.
Engine manufacturers have spent years working around the limitations of the camshaft. The results are myriad proprietary solutions. Honda has VTEC, short for Variable Valve Timing and Lift Electronic Control. Toyota has VVT-i. BMW has VANOS, Ford has VCT. All these systems provide ways to adjust the valve action to some degree. VANOS works by allowing the camshaft to slightly rotate a few degrees relative to its normal timing, similar to moving a tooth or two on the timing chain. While these systems do work, they tend to be mechanically complex, and expensive to repair.
The simple solution would be to go with a camless engine. This would mean eliminating the camshaft, timing belt, and most of the associated hardware. Solenoids or hydraulic actuators open and close the valves in an infinitely variable number of ways. Valves can even be held open indefinitely, effectively shutting down a cylinder when max power isn’t necessary.
So why aren’t we all driving camless engines? There are a few reasons. The advantages of camless engines to camshaft engines are analogous to the advantages of electronic fuel injection (EFI) vs carburetors. At the core, a fuel injector is a solenoid controlled valve. The fuel pump provides constant pressure. The engine control unit (ECU) fires the injectors at just the right time to inject fuel into the cylinders.The computer also leaves the valves open long enough so that the right amount of fuel is injected for the current throttle position. Electronically this is very similar to what would be required for a camless engine. So what gives?
Hackers in their 30’s and beyond will remember that until the late 1970’s and early 1980’s, the carburetor was king. Companies had been experimenting with EFI since the 1950’s. The system didn’t become mainstream until the stiff pollution laws of the 70’s came into effect. Making a clean, fuel-efficient carbureted engine was possible, but there were so many mechanical and electronic actuators required that the EFI was a better alternative. So the laws of the 70’s effectively regulated carburetors out of existence. We’re looking at much the same thing with camless engines. What’s missing are the regulations to force the issue.
All the big manufacturers have experimented with the camless concept. The best effort to date has been from Freevalve, a subsidiary of Koenigsegg. They have a prototype engine running in a Saab. LaunchPoint Technologies have uploaded videos showing some impressive actuator designs LaunchPoint is working with voice coils, the same technology which moves the heads in your hard drive.
None of this means that you can’t have a camless engine now – companies like Wärtsilä and Man have engines commercially available. However, these are giant diesel engines used to drive large ships or generate power. Not exactly what you’d want to put in a your subcompact car! For the hacker set, the best way to get your hands on a camless engine today is to hack one yourself.
Ladies and gentlemen, start hack your engines!
Simple, single-cylinder camless engines are relatively easy to build. Start with a four stroke overhead valve engine from a snowblower, scooter, or the like. Make sure the engine is a non-interference model. This means that it is physically impossible for the valves to crash into the pistons. Add a power source and some solenoids. From there it’s just a matter of creating a control system. Examples are all over the internet. [Sukhjit Singh Banga] built this engine as part of a college project. The control system is a mechanical wheel with electric contacts, similar to a distributor cap and rotor system. [bbaldwin1987’s] Camless Engine Capstone project at West Virginia University uses a microcontroller to operate the solenoids. Note that this project uses two solenoids – one to open and one to close the valve. The engine doesn’t need to rely on a spring for closure. [Brian Miller] also built a camless engine for college, in this case Brigham Young University Idaho Camless Engine. [Brian’s] engine uses hall effect sensors on the original camshaft to fire the solenoids. This route is an excellent stepping stone before making the jump to full electronic control.
It wouldn’t take much work to expand these projects to a multi-cylinder engine. All we’re waiting for is the right hacker to take up the challenge!
[amazingdiyprojects] has been working on a 3D printable jet engine. You may remember seeing a 3D printed jet engine grace our front page back in October. That one was beautiful didn’t function. This one flips those values around. [amazingdiyprojects] seems to make a living from selling plans for his projects, so naturally most of the details of the build are hidden from us. But from what we can see in the video clips there are some really interesting solutions here.
Some of the parts appear to be hand-formed sheet metal. Others are vitamins like bearings and an electric starter. We really liked the starter mechanism, pressing in the motor to engage with a spline, or perhaps by friction, to give the starting rotation.
What really caught our attention was casting the hot parts of the printer in refractory cement using a 3D printed mold. It reminds us of the concrete lathes from World War 1. We wonder what other things could be built using this method? Flame nozzles for a foundry? A concrete tea-kettle. It’s pretty cool.
We’re interested to see how the jet engine performs and how others will improve on the concept. Video of it in action after the break.
The RepRap wiki was spammed this week. Everything is fine now, but I feel I should call attention to the fact that the RepRap wiki needs some people to contribute, organize, and maintain everything. The wikis for obscure anime shows are better than the RepRap wiki, so if you’re looking to contribute to an important open source project, there ‘ya go.
The 200cc, 5.5HP, 4-stroke OHV Honda GX200 engine is found in a whole lot of tools, and is a fantastic power plant to build a go-kart around. It also costs about $350. There are clones of this engine available direct from China for about $100. Here’s how you add a turbo to one of these clone engines.
Freescale makes some pretty cool sensors and [Juan Ignacio Cerrudo] figured they needed breakout boards. He has some boards for a low-power three-axis accelerometer, an accelerometer and magnetometer, and a pressure sensor.
[TheBackyardScientist] hung out with some cub scouts a few weekends ago and launched a high altitude balloon over Florida. The payload included a game camera, APRS tracker, GoPro, and a few other bits and bobs. The balloon reached 106,000 feet and landed only a few miles from Cape Canaveral.
Big RC planes – UAVs especially – are a pain to launch. Flying wings above a certain size are just dangerous to launch by hand, and landing gear is heavy and for the most part unnecessary. What’s the next best solution? A trebuchet, of course. It mounts on a car and is able to give a UAV a little bit of altitude and some speed. A pretty good idea that could be easily implemented with some load-bearing PVC pipe.
About a month ago, a search of AliExpress turned up Apple’s A8 CPU. I bought one. Here’s what I got. It’s a stupidly small pitch BGA, and I don’t have a datasheet. What am I going to do with it? Make a non-functioning board with a few ports, resistors, no traces, and the A8 chip planted square in the middle.
[Fabien-Chouteau] submitted his interesting solenoid engine. In an internal combustion, steam, or pneumatic piston engine, the motive force is produced by expanding gas. In [Fabien]’s little engine it is produced by the arm of a hard drive. Solenoid engines are usually just for show, and come in all shapes and sizes. If you want to move something using electricity an axial motor is probably a better bet. But if you want a challenge and a learning experience, this is hard to beat.
[Fabien] had some problems to solve before his motor made its first revolution. Just like a piston engine the timing needed to be exact. The arm firing at the wrong time could cause all sorts of trouble, the equivalent of backfire in a combustion engine. A STM32f4 discovery board was coupled with a Hall-effect sensor and a MOSFET. When the board read that the arm has moved back to the most efficient position for firing it sent a pulse through the coil. Just like a regular engine, getting the timing right makes all the difference. Once [Fabien] got it tuned up his motor could spin around at a steady 3000 rpm.
There is a public menace on the streets of Perth, Australia: two motorized picnic tables. Police are looking for the drivers of the, erm, vehicles, which were seen cruising down the West Coast Highway, complete with passenger casually sitting on the tables having a drink. It looks like the two picnic tables are being driven by a centrally mounted lawnmower motor and are controlled from one of the seats. The police are interested because it is illegal to drive something on the public road in Australia that isn’t properly licensed. I’ll bet that his insurance probably doesn’t cover taking it out for a spin on the highway, either. The accompanying TV news report does not identify the person responsible, but they claim have spoken to the builder, who says that the two tables can manage up to 50 miles per hour. They claim that he is even working on an upgraded model that includes a built-in barbecue, which could bring a whole new meaning to the term drive-thru.
[RimstarOrg] has posted an awesome writeup on his Hero’s steam engine . Hero’s engine is a Greek design from the first century and is the earliest known steam engine. It’s amazing to think he developed the engine seventeen centuries before the industrial revolution, and yet it was largely ignored. While you can find more faithful replicas, of this landmark machine [RimstarOrg]’s rig can’t be beaten for simplicity and he does a great job of explaining the principal of operation and construction.
Using a soda can filled with water and a propane torch [RimstarOrg] was able to get the can to rotate rapidly by ejecting steam from two holes in the side of the can. A fishing swivel is used to provide a pivoting joint and allow the can to rotate freely.
While we’ve covered steam engines before, we loved this simple design, and can’t wait to see what [RimStarOrg] comes up with next.